Update app.py

app.py

@@ -6,28 +6,37 @@ from PIL import Image, ImageDraw, ImageFont
 import scipy.io.wavfile
 from TTS.api import TTS  # Coqui TTS (open source)
 from moviepy.editor import CompositeVideoClip, ImageClip, AudioFileClip, concatenate_videoclips
+import os
+from groq import Groq
 
-# Use GPT-2 Medium (lighter version) for text generation
-script_generator = pipeline("text-generation", model="gpt2-medium", truncation=True, max_length=100)
+# Load and Initialize Models
+# Use Groq for text generation
+client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
 
 # Use Stable Diffusion (open-source) for image generation
 image_generator = StableDiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-2-1-base", torch_dtype=torch.float16).to("cpu")
 
-# Use MusicGen Small (open-source) for music generation
-music_generator = pipeline("text-to-audio", model="facebook/musicgen-small", device="cpu")
-
 # Use Coqui TTS (open-source) for text-to-speech
 tts = TTS(model_name="tts_models/en/ljspeech/tacotron2-DDC", progress_bar=False, gpu=False)
 
+# Generate Comedy Script
 def generate_comedy_script(prompt):
-    script = script_generator(prompt)[0]['generated_text']
+    response = client.chat.completions.create(
+        messages=[
+            {"role": "user", "content": prompt}
+        ],
+        model="llama-3.1"
+    )
+    script = response["choices"][0]["message"]["content"]
     return script
 
+# Convert Text to Speech using Coqui TTS
 def text_to_speech(script):
     output_audio = 'output.wav'
     tts.tts_to_file(text=script, file_path=output_audio)
     return output_audio
 
+# Create Images Using Stable Diffusion
 def create_images_from_script(script):
     lines = script.split('. ')
     image_paths = []
@@ -38,19 +47,15 @@ def create_images_from_script(script):
         image_paths.append(img_path)
     return image_paths
 
+# Generate Fun Music Track using Groq API
 def generate_fun_music(prompt, output_music_file="fun_music.wav"):
-    # Generate music based on the prompt using MusicGen
-    response = music_generator(prompt)
-    
-    # Extract audio and sampling rate from the response
+    response = client.music.generate(prompt=prompt)
     audio_data = response["audio"]
     sampling_rate = response["sampling_rate"]
-    
-    # Save the generated music to a file
     scipy.io.wavfile.write(output_music_file, rate=sampling_rate, data=audio_data)
-    
     return output_music_file
 
+# Create Video from Generated Images
 def generate_text_video(script):
     image_paths = create_images_from_script(script)
     clips = []
@@ -61,6 +66,7 @@ def generate_text_video(script):
     final_video.write_videofile("/tmp/final_video.mp4", fps=24)
     return "/tmp/final_video.mp4"
 
+# Combine Audio and Video
 def combine_audio_video(video_path, audio_path):
     video = VideoFileClip(video_path)
     audio = AudioFileClip(audio_path)
@@ -68,6 +74,7 @@ def combine_audio_video(video_path, audio_path):
     final_video.write_videofile("/tmp/final_comedy_video.mp4", fps=24)
     return "/tmp/final_comedy_video.mp4"
 
+# Main Function to Generate Comedy Animation
 def generate_comedy_and_animation(prompt):
     script = generate_comedy_script(prompt)
     audio_file = text_to_speech(script)
@@ -76,7 +83,8 @@ def generate_comedy_and_animation(prompt):
     final_video = combine_audio_video(video_file, fun_music)
     return script, audio_file, final_video
 
-def generate_kids_content(theme):
+# Generate Kids Music Animation
+def generate_kids_animation_with_music(theme, output_video_file="kids_animation.mp4"):
     music_file = generate_fun_music(theme, output_music_file="kids_music.wav")
     clips = []
     for i in range(5):
@@ -89,9 +97,15 @@ def generate_kids_content(theme):
         clips.append(ImageClip(frame_path).set_duration(1).set_position(('center', 'center')))
     final_video = CompositeVideoClip(clips, size=(800, 400))
     final_video = final_video.set_audio(AudioFileClip(music_file))
-    final_video.write_videofile("/tmp/kids_animation.mp4", fps=24)
-    return music_file, "/tmp/kids_animation.mp4"
+    final_video.write_videofile(output_video_file, fps=24)
+    return music_file, output_video_file
 
+# Main Function to Generate Kids Content
+def generate_kids_content(theme):
+    music_file, video_file = generate_kids_animation_with_music(theme)
+    return music_file, video_file
+
+# Gradio Interface
 with gr.Blocks() as app:
     gr.Markdown("## AI Comedy and Kids Content Generator")
 
@@ -130,3 +144,4 @@ app.launch()
 
 
 
+